Grafana Cloud on a paid plan can cost well over $30 a month, even before you connect your first data source, because pricing depends on the number of metric series and their retention period. Grafana OSS is exactly the same charting and alerting engine, open source, which you install on your own Box, once, without a monthly bill depending on how many metrics you collect. Below is a step-by-step guide on how to set up Grafana on Saplo.
Why Grafana on your own Box instead of Grafana Cloud
Paid Grafana Cloud plans have two characteristics that are particularly painful when monitoring many projects at once:
- Cost increasing with the number of metrics and their retention time - the more servers and applications you monitor, and the longer you want to keep the history, the higher the monthly bill.
- Free plan limit that will eventually be exceeded - the free tier is enough for one small project, not for several clients at once.
Grafana OSS is the same engine that you set up on your own Box. Zero fees dependent on the number of metrics, telemetry data stays on your infrastructure, and the look and logic of building dashboards are identical to the cloud version.
Requirements
To set up Grafana on Saplo, you need:
- A Box with Docker running (the engine automatically enables it during installation from the catalog, nesting in the container is required, which Saplo has by default),
- At least 512 MB of free RAM on the Box and 2 GB of disk space for the internal database,
- A domain or subdomain
*.saploapp.plincluded in the Price of every plan.
It's important to distinguish right away what Grafana is and isn't. Grafana itself doesn't collect any metrics; it's solely an engine for charts, dashboards, and alerts. You manually connect data sources, such as Prometheus collecting metrics from servers, InfluxDB with time-series data, or a regular MySQL or Postgres database with tables for analysis, after the first launch, in the Connections tab. Grafana on Saplo stores its own data, such as saved dashboards, user accounts, and source configurations, in a separate SQLite database within the data volume, so you don't need any external database for the installation itself.
Step-by-step installation
- In the Saplo panel, go to App catalog and select Grafana (under "Data").
- Indicate the Box where it will be installed and the domain or subdomain under which the application will be available.
- Confirm the installation. The engine deploys a Docker container with the official
grafana/grafana-ossimage, prepares a data volume, generates a random administrator account, and exposes the panel under the specified domain using NGINX with free SSL. - The engine waits for Grafana to actually respond to a health check before considering the installation complete. So, if the image fails to launch, you'll see an error instead of a dead domain.
- Once finished, you'll find the
GRAFANA-SETUP.txtfile in the Box's app directory with the login and generated password for the panel.
First configuration
Grafana, just like MinIO or Vaultwarden, automatically generates an administrator account upon installation instead of asking for it in a wizard.
- Log in to the panel with the credentials from the
GRAFANA-SETUP.txtfile and change the password in Administration -> Users before doing anything else, especially if you are sharing access with a team. - Add the first data source in Connections -> Data sources. Grafana supports Prometheus, InfluxDB, MySQL, PostgreSQL, Elasticsearch, and Loki, among others, and you can find a list of supported sources in the official Grafana documentation.
- After connecting the source, build your first dashboard in Dashboards -> New, or import a ready-made template from the Grafana gallery if you are monitoring a popular service, such as an nginx server or a Postgres database.
- Configure alerts in Alerting if you want to receive notifications when a metric exceeds a set threshold, for example, disk usage above 90 percent.
Grafana on Saplo vs. Grafana Cloud
| Grafana Cloud Pro | Grafana on Saplo | |
|---|---|---|
| Price | increases with the number of metric series and retention time | included in your Saplo plan, no extra charge for metrics |
| Where is the data | Grafana Labs infrastructure | Your Box |
| Engine version | the same code as Grafana OSS plus cloud features | Grafana OSS, official grafana/grafana-oss image |
| Data sources | the same plugins and integrations | the same plugins and integrations |
| Who administers | Grafana Labs | You, but without manually setting up the server |
FAQ
Does Grafana collect metrics on its own, or do I need something else? Grafana itself doesn't collect anything; it's purely a charting and alerting engine. Metrics are provided by a separate data source, such as Prometheus or an SQL database, which you connect manually after installation in the Connections tab.
What data sources can I connect? Grafana supports, among others, Prometheus, InfluxDB, MySQL, PostgreSQL, Elasticsearch, and Loki. You can find a full list of plugins in the panel when adding a new data source.
Can I monitor the Box where Grafana is hosted? Yes, but you'll need a separate application to collect system metrics, such as Prometheus with an appropriate exporter, and connect it as a data source. The Grafana installation itself does not collect such metrics.
Are data in Grafana automatically backed up? This depends on your Box plan. Backup works at the Box level, just like with other applications from the App catalog; it's not a separate function of Grafana itself.
Can I have more than one dashboard and multiple users? Yes, the number of dashboards and user accounts is not limited by Grafana OSS; the only limit is the resources of your Box plan.
If you have any questions about setting up Grafana on your own Box or connecting your first data source, contact us, and we'll help you with the initial configuration.
- #Dashboards
- #Grafana
- #Metrics
- #Monitoring
- #Guide
